1. Implementation of a 2 input AND and OR logic function using perceptron. Start with different set of initial weights and show that there are more than one solution to the problem.
2. Develop an algorithm using Hebbian Learning rule to solve NAND and NOR problem. Assume the elements of initial weight matrix as random value and study the effect of different learning rates.
3. Solve the following Boolean function using Perceptron learning rule:
Y= x1´ x2´ x3´ + x1´ x2´ x3+ x1 x2´ x3´+ x1 x2 x3
4. Consider the following function:
F(x) = x13 + 2x1x2 - x12 x22
Perform two iterations of Newton’s method from the initial guess x0 = [1 -1]T.
5. Demonstrate that EX-OR and XNOR gate is a non-linearly separable problem. Design a MLP for the purpose and train it using BP algorithm. Assume the use of a logistic function for the nonlinearity.
6. Given the dataset {-2,-1,0,1,2} with targets t(-2)=0, t(-1)=0.25, t(0)=0.5, t(1)=0.75 and t(2)=1. Determine the weights of all neurons with sigmoid transfer function such that the MSE is almost zero and if we use the neural network with five neurons in the first layer and one neuron in the second layer.
7. Using a MLP with BP algorithm approximate the function y=1+Sin(2px)for -1=x=1. Draw the actual output and simulated output. Show the signal matching after 100/ 500/ 1000 iterations.
8. Implement a 2-input EX-OR gate using regularized RBF (use 4 centres). Plot the output surface for different input variables. Extend this for a 3 input RBF using 8 centers. Tabulate the input output pattern mapping for each case.
9. Using Mamdani and Sugeno model Fuzzy system, design a non-stationary time series prediction network to predict the output against time sample.
10. Minor Project

To design and development of soft computing algorithms for solution of real time problem.

To develop and model the real time software using soft computing techniques

A student will be to learn how to develop the different Soft Computing techniques, e.g. neural networks, fuzzy logic using MATLAB/C platform for real time application.

S. Haykin, Neural Networks - A Comprehensive Foundation, Pearson Education, India

Jang, Sun and Mizutani, Neuro-Fuzzy and Soft-Computing – A computational approach to learning and machine intelligence, Prentice Hall of India

Satish Kumar, Neural Networks: A Classroom approach, Tata Mcgraw Hill

Martin T. Hagan, Howard B. Demuth, Mark H. Beale, Neural Network Design, Thomson